Roadmap for the management of 3 million digital objects. Towards a certified repository

Editor's Notes

• #2: I am Saskia van Bergen, I am working as a project manager for the department Innovation and Projects of Leiden University Libraries. My projects mainly concern digital special collections or digital heritage. https://www.universiteitleiden.nl/en/staffmembers/saskia-van-bergen#tab-1
• #3: Facilitating availability & use of digital special collections form important tasks of our library. We digitize our own paper heritage, both in-house and externally, and participate in various digitization programs. Previous, the digital heritage collections could be found in different systems using a variety of ways to present and store them. This was not very practical for our users, but also a lot of work and expensive to maintain. For this reason, the library started a project to redesign the infrastructure about 4 years ago. The library has built its own infrastructure for the management of digital heritage, based on Islandora. At the moment our repository contains about 1,5 million digital objects (40 TB all together). We have nearly finished the migration of our digital heritage collections from the old infrastructure to Islandora, but we also have digitized a lot of collections that haven’t been ingested yet because of our migration project, and we are also starting new projects resulting in digital collections as well.   In 2019 we plan to migrate our scholarly repository (containing the publications of Leiden university) to the Islandora environment. In the course of 2019, our complete digital collection is thus expected to grow up to 3 million objects.
• #4: Islandora is OAIS compliant, but this is not a standard for quality of service; it is a reference model that operates at a conceptual level. Institutions began to declare themselves ‘OAIS-compliant’ to underscore the trustworthiness of their digital repositories. However, claims of trustworthiness are easy to make but are far more difficult to justify or objectively prove. Merely storing the data in a repository is not sufficient to guarantee future access and reuse. This presentation focuses on the roadmap for improving the visibility, usability, and sustainability of our digital heritage collection (I will leave the scholarly repository out, because a lot will change with the migration from DSpace to Islandora) It contains two parts: First I would like to tell you more about the roadmap itself: what does it consist of, how is it used in our organization? In the second part I want to focus on self-assessment (how do I know I am doing the right thing?) related to the certification of repositories.  
• #5: The entire chain from selection and acquisition to availability and use of digital collections is complex and involves a large number of departments and persons. At the UBL, the responsibilities for the individual steps in the process are generally well allocated: •The Centre for Digital Scholarship identifies needs in the field of digitization and use of digital collections by researchers. •Digitization is done either in-house or outsourced to an authorized digitization company. •The curators are responsible for selection, description and availability, with respect of any copyright restrictions. •The workflows relating to copyright, such as agreements with rights holders and the clearing of rights are responsibility of the Special Collections Department. •Our Metadata Department advises on metadata and standards and supports batch operations •The department Digital Services manages the infrastructure for storage, management and use of the digital collection. •The Project Manager for the special collections coordinates large-scale digitization projects. She is also a consultant in the field of digital sustainability, standards and innovation. This often leads to policy advices
• #6: Decisions regarding policy and implementation are ideally based on a full consideration of the entire chain. The main problem in Leiden was that it was unclear who bore the overall responsibility for the digital collections. Although there were regular meetings between the various departments involved, this was not organized structurally and the meetings were usually limited to a number of departments of the chain. There was a need for a dedicated manager of the digital collection as a whole. In creating a role for this UBL is not unique. In Leiden this role has been named Process Coordinator, but names such as Digital Curator or Archivist, and manager of digital collections can also be found. This role has been (at least for the time being) given to me.   The following tasks have been assigned to the UBL Process Coordinator. I: •have a leading role in setting up the work processes around digital heritage, such as digitization projects, the development of the digitization workflow, infrastructure and the online presentation. •monitor the work processes across the various departments, with explicit attention to their cohesion •am responsible for drawing up agreements, procedures, regulations etc. and ensures that they are complied with, both during and after the end of an activity or project; •am responsible for the communication of agreements, procedures, regulations etc. within the UBL. •am responsible for the administration of the digital collections •draw up policy notes for the Executive Board and I monitor the implementation of decisions taken
• #7: digital collections. Drawing up this roadmap helped to get a more concrete picture of the goals. Each year agreements are made with regard to the progress of the Roadmap. In the first place, for each task a set of essential tools and documents was defined. These consisted for example of policy documents, process descriptions, templates, checklists, spreadsheets with calculations and descriptions of digitization projects.   It turned out that several colleagues in the library had been working on these documents and tools already, but they were never stored in one place, and their status was not always clear. The next step was to analyze the status for each of these, defining: 1. The first steps have yet to be taken 2. Work is in progress, but it is still ongoing 3. It has it definite form, is in production and only has to be updated periodically   On the slide you see some examples of this: Framework Policy regarding open access Overview of preferred formats   Process Sharepoint copyright clearing   Key figures Overview costs of digitization   Consultancy / best practice FAQ on the website (for researchers
• #8: But how did I know I was doing the right thing? How do I measure quality, regarding to the management of the digital collections?   At that point, Certification came in sight.
• #9: In the world, and specifically in the Netherlands, there are more and more organizations that store large quantities of digital material in a sustainable manner: e.g. our National Archive, Institute for Sound & Image, have large digital repositories. Everyone assumes that these institutions work professionally and that the digital collections are stored safely. Nevertheless, this trust will periodically have to be clearly demonstrated to external parties: through audit and certification.
• #11: Recent developments in digital trustworthiness have led to the development of criteria and certifications that can be used as tools when auditing digital preservation systems, like the Audit and Certification of Trustworthy Digital Repositories (ISO 16363). The Core Trust Seal (CTS) method was developed as Data Seal of Approval by the Dutch organization Data Archiving and Networked Services (DANS), in the effort to develop a basic certification distilled from two other ones. The certification is formulated in 16 guidelines for the application and verification of quality aspects with regard to creation, storage and reuse of data, and has a strong focus on research data.  
• #12: A repository must follow to stages to obtain a Core Trust Seal: the first is to perform a self-assessment against the requirements. Afterwards, the self-assessment is peer-reviewed by a member of the CTS assessment group who provides feedback. The organization applying for the certification can then modify and improve the assessment and submit it for a final round of review.
• #13: The developers of the certification list seven main benefits of acquiring the Core Trust Seal: 1) Trustworthiness, 2) Stakeholder confidence, 3) Improvements in processes, 4) Transparency, 5) Differentiation from others, 6) Awareness raising about digital preservation, and 7) Make digital preservation less labor- and time-intensive. While these benefits are valuable for a digital repository, self-assessments and certification can be labor intensive, time consuming, and expensive. By their very nature, assessments like the Core Trust Seal open digital repositories to outside scrutiny, and this scrutiny can be seen as intrusive. Even the more basic level certification can be difficult to achieve if underestimated. Moreover, if embracing standards and undergoing self-assessment can help building both self-confidence (among the staff) and ensure that the organization is following good practices, we may start questioning the need of the certification itself. Is certification a priority for digital repositories? Can a repository be trustworthy without becoming a certified trusted digital repository? Are certification and trust as interrelated as is often assumed?   For this reason we asked an intern to perform a preliminary investigation for the certification of our repository. She examined the benefits and the risks of acquiring the CTS in relation to the Islandora repository of UBL.   Her research consisted of 3 parts: 1 Analysis of requirements: how far are we already? Completely lacking Needs an update (Almost) done   2 Interview Dutch repository managers Responders were asked to discuss: -Reasons for certification process -How they prepared for it -Difficulties encountered -The benefits they experienced from the process -Time investments   3 Time investment vs benefits of stakeholders at UBL Business case analysis: NO not needed YES right now YES but later
• #14: Our intern made a business case analysis of all internal and external stakeholders, who can affect and be affected by the certification. Each of them is different in terms of importance, influence and needs. For example: the influence of Dutch funding agencies is potentially high. Certification signifies to funders that the data they have invested in will continue to be available for reuse. Without certification we risk to miss opportunities of financial support. The use of a certificatied repository is not a requirement for funding yet, but it is expected to be in the near future.  
• #15: Her conclusion: There still a lot to be done before certification is within reach. Although it is tempting to simply start a certification project. Main advice: Start a.s.a.p. but focus on self-assessment in the first place: improvement of the internal processes Use the opportunity of the migration process of the digital collection into the Islandora environment to create procedures to address changes to the repository. Create a list of activities with a schedule, deliverables and objectives. Focus first on the development of a well-formulated preservation plan: general and specific needs for the care of collections are determined, priorities are established, and resources for implementation are identified.   The first 3 bullets of the advice were already started with the roadmap we talked about in the first place, but the last bullet was new to us. It is a very important recommendation, because it is the heart of the whole audit process and offers context to the activities identified in the roadmap.
• #16: For the last couple of years, various data repositories in the world have successfully applied for a certification seal. Most seal holders refer to external benefits in the form of increased awareness of the value of their repositories to their communities and funders, but the experience of UBL is that the internal benefits should not be underestimated either. For our library, it all started with the need to improve the management of an expanding repository. To be able to curate 3 million objects professionalization of administration and documentation is essential. Creating the role of the Process Manager and drawing up a Roadmap functioned as a catalyst for change. At present, we have a clear idea of the requirements, and this helps us to draw up a workplan for the next couple of years. More importantly, starting with a self-assessment, we will also achieve a greater sense of involvement, as everyone will have become more aware of his or her role and responsibility in the whole process. Within this process, certification is not perceived as a goal, but will simply be the result of our activities.